Please Help Formatting Date in Units to MDY

[Bob Honda]

Hello,

I am trying to convert the date in units, for example 613608, into a date variable preferable in the *MYD* format. The units are based on "hours since 1800-01-01 00:00:0.0". So in this case I believe 613608 should result in January 1st, 1870, which is the first date in the dataset. I have tried to look up the syntax for the datetime commands, but I am still very confused.

I have provided the code below with the first 4 observations in my dataset. Any help would be greatly appreciated.

input long time float(lat lon pdsi)
613608 -46.25 168.75    .1647371
613608 -43.75 -73.75    .6625475
613608 -43.75 -71.25    .7405886
613608 -43.75 168.75   .10170344

Thank you!

[Leonardo Guizzetti]

Stata has different types of internal representations of time: dates and datetime (see -help datetime-). Only datetime representations have the necessary precision to deal with units of time smaller than a day, so you must perform any calculation on and using functions for datetime.

The main bit is that since you have number of hours, and datetimes are recorded in units of milliseconds, a simple conversion is needed from hours to ms (= 60*60*1000). This is the offset, to be added to the time of your reference date. The date can be computed using the -clock()- function.

The rest then is just formatting to some desirable display format. Additional formatting can be found at -help datetime display formats-.

In the code below I used a macro to store the conversion factor, but this is a matter of taste and is unnecessary. However, it is good practice to store things like constants in their own variables (or macros, here) so they do not appear as hard-coded "magic" numbers in code.

local hours_to_ms = 60 * 60 * 1000
gen double want = clock("1jan1800", "DMY") + time * `hours_to_ms'
format %tcNNCCYYDD want
list

Result

     +------------------------------------------------+
     |   time      lat      lon       pdsi       want |
     |------------------------------------------------|
  1. | 613608   -46.25   168.75   .1647371   01187001 |
  2. | 613608   -43.75   -73.75   .6625475   01187001 |
  3. | 613608   -43.75   -71.25   .7405886   01187001 |
  4. | 613608   -43.75   168.75   .1017034   01187001 |
     +------------------------------------------------+

[Bob Honda]

Thank you Leonardo it worked perfectly!